High resolution allele identification

Abstract

Provided herein are methods for accurately determining the alleles present at a locus that is broadly applicable to any locus, including highly polymorphic loci such as HLA loci, BGA loci and HV loci. Embodiments of the disclosed methods are useful in a wide range of applications, including, for example, organ transplantation, personalized medicine, diagnostics, forensics and anthropology.

We claim: 
     
       1. A computer-implemented method comprising:
 performing a nucleic acid amplification process that produces an amplification product that comprises a nucleic acid sequence of a subject, wherein the nucleic acid sequence comprises one or more single nucleotide polymorphisms (SNPs); 
 performing a sequencing process on the amplification product that produces a plurality of sequencing reads; 
 mapping the plurality of sequencing reads to a reference sequence to identify pairs of candidate alleles; 
 for each pair of candidate alleles, determining genotype log-likelihood scores for each individual SNP, each genotype log-likelihood score being a sum of log-probabilities for each individual SNP that the pair of candidate alleles could account for sequences present at the individual SNP in the sequencing reads that map to the individual SNP; 
 for each pair of candidate alleles, determining phase log-likelihood scores for each sequential pair of SNPs, each phase log-likelihood score being a sum of log-probabilities for each sequential pair of SNPs that the pair of candidate alleles could account for sequences present at the sequential pair of SNPs in the sequencing reads that map to the sequential pair of SNPs; and 
 selecting a pair of candidate alleles for which a sum of the genotype log-likelihood score and the phase log-likelihood score is highest as alleles present at a locus. 
 
     
     
       2. The method of  claim 1 , wherein the plurality of sequencing reads is comprised of sequencing reads of 35-100 base pairs. 
     
     
       3. The method of  claim 1 , further comprising: for each pair of candidate alleles, determining a frequency log-likelihood score, the frequency log-likelihood score being a sum of log-frequencies at which each of the pair of candidate alleles are present in a human population. 
     
     
       4. The method of  claim 3 , wherein selecting the pair of candidate alleles for which the sum of the genotype log-likelihood score and the phase log-likelihood score is highest as the alleles present at the locus comprises selecting the pair of candidate alleles for which the sum of the genotype log-likelihood score, the phase log-likelihood score, and the frequency log-likelihood score is highest as the alleles present at the locus. 
     
     
       5. The method of  claim 1 , wherein the reference sequence comprises a genome sequence and a plurality of allele sequences, and wherein the genome sequence is a human genome sequence and the plurality of allele sequences are human sequences. 
     
     
       6. The method of  claim 5 , wherein a sequence in the genome sequence has been removed or masked. 
     
     
       7. The method of  claim 5 , wherein the human genome sequence is GRCh37/hg19. 
     
     
       8. The method of  claim 5 , wherein the plurality of allele sequences are selected from a set of protein groups. 
     
     
       9. The method of  claim 5 , wherein the genome sequence comprises transcriptome sequences, whole exome sequences, or whole genome sequences. 
     
     
       10. The method of  claim 1 , wherein mapping further comprises:
 identifying, as a first set of candidate alleles, alleles to which map a greatest number of sequencing reads; 
 identifying, as a second set of candidate alleles, alleles to which map the greatest number of sequencing reads, excluding the sequencing reads that map to the first set of candidate alleles; and 
 if less than 90% of the sequencing reads that map to the locus map to an allele of the first set of candidate alleles or the second set of candidate alleles, identifying, as a third set of candidate alleles, alleles to which map the greatest number of sequencing reads, excluding the sequencing reads that map to the first set of candidate alleles or the second set of candidate alleles. 
 
     
     
       11. The method of  claim 10 , wherein, if the number of sequencing reads that map to the locus following exclusion of the sequencing reads that map to the first set of candidate alleles is greater than 1% of the number of sequencing reads that map to the first set of candidate alleles, further identifying, as a subset of the second set of candidate alleles, alleles to which map a second greatest number of sequencing reads without excluding the sequencing reads that map to the first set of candidate alleles. 
     
     
       12. The method of  claim 10 , wherein the third set of candidate alleles are only identified if the number of sequencing reads that map to the alleles to which map the greatest number of sequencing reads, excluding the sequencing reads that map to the first set of candidate alleles or the second set of candidate alleles, make up at least 10% of a total number of sequencing reads that map to the locus. 
     
     
       13. The method of  claim 1 , further comprising receiving sequence data, the sequence data comprising the plurality of sequencing reads, wherein the sequence data comprises genome-wide sequencing data. 
     
     
       14. The method of  claim 13 , wherein the genome-wide sequencing data are transcriptome sequencing data, whole exome sequencing data, or whole genome sequencing data. 
     
     
       15. The method of  claim 13 , wherein coverage of the sequence data is at least 30 fold. 
     
     
       16. The method of  claim 13 , wherein coverage of the sequence data ranges from 30 fold to 100 fold, and wherein the plurality of sequencing reads are from DNA. 
     
     
       17. The method of  claim 13 , wherein coverage of the sequence data ranges from 100 fold to 500 fold, and wherein the plurality of sequencing reads are from RNA. 
     
     
       18. The method of  claim 13 , wherein coverage of the sequence data is at least 1000 fold, and the plurality of sequencing reads are from a targeted sequence. 
     
     
       19. The method of  claim 1 , wherein the plurality of sequencing reads are paired-end reads. 
     
     
       20. The method of  claim 1 , wherein the plurality of sequencing reads are single-end reads. 
     
     
       21. The method of  claim 1 , wherein the locus is a highly polymorphic locus. 
     
     
       22. The method of  claim 1 , wherein the alleles present at the locus comprise an HLA type at the locus. 
     
     
       23. The method of  claim 22 , wherein a sample is taken from a cell to be transplanted into the subject, and further comprising determining if an HLA type of the cell matches the HLA type at the locus. 
     
     
       24. The method of  claim 23 , further comprising transplanting the cell into the subject if the HLA type of the cell matches the HLA type at the locus. 
     
     
       25. The method of  claim 23 , further comprising:
 transplanting the cell into the subject if the HLA type of the cell matches the HLA type at the locus; and 
 administering an agent that reduces a likelihood of transplant rejection to the subject. 
 
     
     
       26. The method of  claim 1  further comprising, determining, based on the alleles present at the locus, a geographic origin of the subject. 
     
     
       27. The method of  claim 1  further comprising, linking, based on the alleles present at the locus, the subject to an individual.